U.S. patent number 7,177,515 [Application Number 10/140,535] was granted by the patent office on 2007-02-13 for surface plasmon devices.
This patent grant is currently assigned to The Regents of the University of Colorado. Invention is credited to Michael J. Estes, Garret Moddel.
United States Patent |
7,177,515 |
Estes , et al. |
February 13, 2007 |
Surface plasmon devices
Abstract
A device including an input port configured to receive an input
signal is described. The device also includes an output port and a
structure, which structure includes a tunneling junction connected
with the input port and the output port. The tunneling junction is
configured in a way (i) which provides electrons in a particular
energy state within the structure, (ii) which produces surface
plasmons in response to the input signal, (iii) which causes the
structure to act as a waveguide for directing at least a portion of
the surface plasmons along a predetermined path toward the output
port such that the surface plasmons so directed interact with the
electrons in a particular way, and (iv) which produces at the
output port an output signal resulting from the particular
interaction between the electrons and the surface plasmons.
Inventors: |
Estes; Michael J. (Longmont,
CO), Moddel; Garret (Boulder, CO) |
Assignee: |
The Regents of the University of
Colorado (Boulder, CO)
|
Family
ID: |
29269688 |
Appl.
No.: |
10/140,535 |
Filed: |
May 6, 2002 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20030206708 A1 |
Nov 6, 2003 |
|
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
10103054 |
Mar 20, 2002 |
7010183 |
|
|
|
Current U.S.
Class: |
385/130; 385/30;
385/31 |
Current CPC
Class: |
B82Y
20/00 (20130101); G02B 6/10 (20130101); G02B
6/12 (20130101); G02B 6/1226 (20130101) |
Current International
Class: |
G02B
6/10 (20060101); G02B 6/26 (20060101) |
Field of
Search: |
;385/129-132,30,31,37 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
Other References
N A. Janunts et al, Modulation of Light Radiation during Input into
Waveguide by Resonance Excitation of Surface Plasmons, Jul. 16,
2001, Applied Physics Letters, V 79, No. 3, p. 299-301. cited by
other .
N. A. Janunts et al, TASERs: Possible DC Pumped Terahertz Lasers
Using Interwell Transitions in Semiconductor Heterostructures, Oct.
10, 1994, Applied Physics Letters, V 65, No. 15, p. 1865-1867.
cited by other .
H. Drexler et al. Photon-Assisted Tunneling in a Resonant Tunneling
Diode: Stimulated Emission and Absorption in the THz Range, Nov. 6,
1995, Applied Physics Letters, V 67, No. 19, pp. 4102-4104. cited
by other .
K. Kempa et al, Towards Stimulated Generation of Coherent Plasmons
in Nanostructures, Mar. 1, 1999, Journal of Applied Physics, V 85,
No. 7, pp. 3708-3712. cited by other .
M. Asada et al., Estimation of Interwell Terahertz Gain by
Photon-assisted Tunneling Measurement in Triple-Barrier Resonant
Tunneling Diodes, Jul. 31, 2000, Applied Physics Letters, V 77, No.
5, pp. 618-620. cited by other .
A. Tredicucci et al, Surface Plasmon Quantum Cascade Lasers at
.lamda..about..mu.m, Oct. 9, 2000, Applied Physics Letters, V 77,
No. 15, pp. 2286-2288. cited by other .
R. Volkov et al, Tunneling-Assisted Photon Emission in MIM
Junctions, Jun. 1991, Physics Stat Sol (b) 163.311. cited by other
.
A Aleksanyan et al, Feasibility of Developing a Tunable Oscillator
Utilizing a System of Metal-Barrier-Metal-Barrier-Metal Junctions,
May 11, 1981, Sov J Quan Elec, pp. 635-637. cited by other .
E. Belenov et al, Investigation of the Radiation Emitted by
Metal-Barrier-Metal Structures, Apr. 13, 1983, Sov Journal Quantum
Electron, pp. 451-455. cited by other .
D. Siu et al, Stimulated Electron Tunneling in Metal-Barrier-Metal
Structures due to Surface Plasmons, Apr. 1, 1976, Applied Physics
Letters, V 28, No. 7, pp. 407-410. cited by other .
D. Drury et al, Theory of Infrared and Optical Frequency
Amplification in Metal-Barrier-Metal Diodes, Jun. 1979, IEE Trans
on Microwave Th and Tech, V MTT-27, No. 6, pp. 598-603. cited by
other .
D. Drury et al, A Stimulated Inelastic Tunneling Theory of Negative
Differential Resistance in Metal-Insulator-Metal Diodes, Jan. 1980,
IEEE J of Quan Elec, V QE-16, No. 1, pp. 58-69. cited by other
.
E. Belenov et al., Resonant Tunneling in Multilayer Structures in
the Presence of Surface Electromagnetic Waves, Apr. 12, 1986, Sov
Tech Phys Lett, pp. 200-202. cited by other .
E. Belenov et al., Emission of Surface Electromagnetic Waves in the
Case of Resonance Tunneling on Electrons, Oct. 17, 1987, Sov J
Quantum Electron, pp. 1348-1352. cited by other .
E. Belenov et al, Amplification of Surface Plasma Oscillations in
Complex Metal-Barrier-Metal Structures, Jul. 12, 1982, Sov J
Quantum Electron, pp. 930-931. cited by other .
L Xie, Stimulated Emission of Surface Plasmons in
Metal-Insulator-Metal (Transition Metal Type Like) Structures, May
1985, Infrared Phys, V 25, No. 5, pp. 661-664. cited by other .
J. Schildraut, Lon-Range Surface Plasmon Electrooptic Modulator,
Nov. 1, 1998, Aplied Optics, V 27, No. 21, pp. 4587-4590. cited by
other .
Levi "Optical Interconnects in systems," Proceedings of the IEEE,
vol. 88, pp. 750-757, 2000. cited by other .
Markoff, "New material to allow plastic transistors," New York
Times, Dec. 3, 2002. cited by other .
Paulson, "Researchers bring wireless communications to the chip,"
IEEE Computer, Sep. 2002. cited by other .
Savage, "Linking with light," IEEE Spectrum, 39(8), pp. 32-36, Aug.
2002. cited by other .
Soller et al., "Energy transfer at optical frequencies to
silicon-based waveguiding structures," J. Opt. Soc. Am. A, 18(10),
pp. 2577-2584 (2001). cited by other .
Soole et al., "Mode-selective detection of optical guided waves by
integrated metal-oxide-metal structures," Opt. Lett., 12(7), pp.
536-538, Jul. 1987. cited by other.
|
Primary Examiner: Song; Sarah
Attorney, Agent or Firm: Pritzkau Patent Group, LLC
Parent Case Text
RELATED APPLICATIONS
The present application is a Continuation-In-Part of U.S. patent
application Ser. No. 10/103,054 filed Mar. 20, 2002, now U.S. Pat.
No. 7,010,183 which application is included herein by reference in
its entirety.
Claims
What is claimed is:
1. A device comprising: a) an input port configured to receive an
input signal, and b) a structure including an elongated tunneling
junction connected with said input port configured such that said
input signal produces a traveling wave along a predetermined path
within said elongated tunneling junction.
2. A device comprising: a) an input port configured to receive an
input signal, and b) a structure including an elongated tunneling
junction connected with said input port configured such that said
input signal enters said elongated tunneling junction from only one
end so as to produce a traveling wave along a predetermined path
within said elongated tunneling junction.
3. The device of claim 2 wherein said input port is an antenna
structure.
4. A device comprising: a) a structure including an elongated
tunneling junction that transmits a traveling wave signal that
travels along a predetermined path within said tunneling junction
and exits said elongated tunneling junction from only one end of
the elongated tunneling junction, and b) an output port connected
with said one end of said elongated tunneling junction.
5. A device comprising: an input port configured to receive an
input signal; and a structure including a tunneling junction
connected with said input port configured such that said input
signal enters said tunneling junction from only one end so as to
produce a traveling wave along a predetermined path within said
tunneling junction, wherein said tunneling junction is configured
in a way which (i) provides electrons in a particular energy state
within said structure and (ii) produces surface plasmons in
response to said input signal.
6. A device comprising: an input port configured to receive an
input signal; and a structure including a tunneling junction
connected with said input port configured such that said input
signal enters said tunneling junction from only one end so as to
produce a traveling wave along a predetermined path within said
tunneling junction, wherein said tunneling junction is configured
in a way which (i) provides electrons in a particular energy state
within said structure and (ii) produces surface plasmons in
response to said input signal, and said device further comprising
an output port, and wherein said tunneling junction is further
configured in a way which (iii) causes said structure to act as a
waveguide for directing at least a portion of said surface plasmons
along said predetermined path toward said output port such that the
surface plasmons interact with said electrons in a particular way,
and (iv) produces at said output port an output signal resulting
from said particular interaction between said electrons and said
surface plasmons.
Description
BACKGROUND OF THE INVENTION
The present invention relates to surface plasmon devices and
methods of making such devices and, more specifically, to surface
plasmon guided wave devices whose optical frequency response is not
limited by RC time constant.
One of the disadvantages of conventional electronic devices based
on electron charge oscillation is that they cannot operate at
terahertz frequencies or higher due to time constant limitations.
In tunneling junction devices, for example, the speed of the device
is limited by the fact that the outer conductive layers of the
tunneling junctions act as parallel plate capacitors. Coupled with
the internal and external resistance to current flow, the RC time
constant constrains the speed of the tunneling junction device.
Thus, only very small devices, which have correspondingly small
capacitance, can be operated at optical frequencies of 10.sup.13
10.sup.15 Hz.
One way to circumvent the RC time constant limitation and get
optical frequency response from these devices is to use traveling
waves. In metal-insulator based structures, such as
metal-insulator-metal tunnel junctions, surface plasmons provide
such a solution. Surface plasmons are TM electromagnetic modes
which propagate along the interface between two materials having
dielectric constants of opposite signs, such as the interface
between a metal and a dielectric material or the interface between
a metal and a semiconductor. The use of surface plasmons as a means
of transferring and manipulating electromagnetic energy in a device
is advantageous because the electromagnetic energy is tightly
confined when in the form of surface plasmons. Therefore, small,
compact devices for controlling electromagnetic energy are enabled
by taking advantage of the properties of surface plasmon
propagation.
Research into devices based on surface plasmon transmission has
focused on devices formed of semiconductor materials and on devices
formed of metal-insulator combinations. Studies on semiconductor
structures have centered mainly on resonant tunneling diode
structures based on multiple quantum wells. For example, Korotkov
et al. examine the theoretical possibility of lasing from double
quantum well (or triple-barrier) heterostructures..sup.1 Korotkov
et al. describe the use of radiative, interwell transitions (i.e.,
inelastic tunneling between adjacent quantum wells) to achieve
lasing. Korotkov et al. also suggest the minimization of absorption
of terahertz radiation in doped gallium-arsenide by using
periodically-spaced p+ and n+ doped injection layers.
Whereas the paper of Korotkov et al. is based on theory, Drexler et
al. discuss experimental behavior of a semiconductor triple-barrier
resonant tunneling diode (RTD) with a bowtie antenna..sup.2 Drexler
et al. observed evidence of stimulated absorption and emission of
terahertz fields in the I-V curve of their experimental device,
indicating photon-assisted tunneling (which they interpret as
stimulated absorption of photons) in the RTD. Drexler et al.
further suggest that the device may be adopted to design tunable
quantum detectors and radiation sources in the terahertz radiation
range.
An approach toward attaining gain in solid state plasmas in
semiconductor structures is discussed by Kempa et al..sup.3 The
article of Kempa et al. discusses the possibility of attaining
growing coherent plasma oscillations (i.e., plasma instability or
stimulated emission of plasmons) using quantum well structures with
the aim of fabricating a terahertz source based on the resulting
gain. By interpreting the experimental and theoretical nonlinearity
of the I V curve of their semiconductor devices, Kempa et al.
assert the presence of population inversion and predict the
achievement of stimulated emission of plasmons using their
semiconductor structure scheme if the structure is optimized.
Estimated gain in a triple barrier (i.e., double quantum well) RTD
structure is discussed by Asada et al..sup.4 By separating the
photon absorption and stimulated emission peaks from the total
current curve in the I-V characteristics of the device, Asada et
al. estimate the net stimulated emission rate of the tunneling
electrons as a function of bias voltage. Thus, Asada et al.
calculate the estimated intersubband terahertz gain and predict
that high gain is possible in the device by increasing the current
density in the device as well as by using thin barriers.
A surface plasmon quantum cascade (QC) laser based on chirped
semiconductor superlattices has been demonstrated by Tredicucci et
al..sup.5 The surface plasmon QC laser of Tredicucci operates at a
wavelength of .lamda..about.19 .mu.m, and relies on guided surface
plasmon modes at a metal-semiconductor interface to effectively
confine the infrared wavelength light. The surface plasmon QC laser
is operable up to a maximum temperature of 145K.
Semiconductor-based devices have several drawbacks. Due to their
crystalline nature and the fact that deposited semiconductor layers
must epitaxial, semiconductor-based devices are relatively
expensive to fabricate and very difficult to integrate with other
optoelectronic devices. In addition, semiconductor carrier
concentration is limited to <10.sup.20 cm.sup.-3, which is about
1000 times lower than in metals. One consequence of the lower
carrier concentration is that the plasma frequency is reduced from
optical frequencies to millimeter wave frequencies, so that optical
waves cannot be carried by semiconductor surface plasmons. Another
consequence is that the device bandwidth is reduced due to
transport resistance. Still another consequence for amplification
and emission devices is that the increased resistance would limit
the output intensity and increase heating.
An alternative to semiconductor-metal structures is the use of a
metal-insulator interface to confine the surface plasmons.
Metal-insulator layers are relatively inexpensive to manufacture in
comparison to semiconductor-based systems. As the oxides of metals
are generally used as the insulator materials, structures including
metal-insulator interfaces are straightforward to fabricate
compared to the epitaxial growth techniques for semiconductor
layers. Due to the high carrier concentration in metals and the
extremely short transit time of tunneling electrons,
metal-insulator-metal (MIM) tunneling junctions are potentially
useful in devices for optical frequency electronics.
Research on surface plasmons in MIM devices published in the 1970's
and 1980's focused largely on developing an understanding the
properties of point contact tunneling junctions, which were shown
to detect infrared and millimeter wave radiation, and on
light-emitting tunneling junctions, which give off a broad spectrum
of spontaneously-emitted visible and infrared radiation.
An emitter based on the MIM diode structure is disclosed by Lambe
et al. in U.S. Pat. No. 4,163,920..sup.6 The emitter of Lambe
includes a strip of metal, with an oxide formed thereon, and a
second metal strip crossing the first metal strip and oxide, thus
forming an MIM structure at the intersection of the two metal
strips. The top surface of the second metal strip is roughened such
that the second metal strip is rendered slightly porous. By
applying a current across the two metal strips, Lambe et al.
achieve light emission from the excitation of surface plasmons in
the tunneling junction due to inelastic tunneling of electrons. In
other words, the emitter of Lambe et al. is essentially a broadband
emitter in the visible spectrum with the light being emitted from
the roughened surface of the MIM junction.
The emitter as described in Lambe et al. has a number of drawbacks.
One such drawback is the relatively inefficient spontaneous
emission process of surface plasmons in single-insulator MIM
diodes. For example, Volkov and Chuiko estimate the efficiency of
this transition to be .about.10.sup.-4..sup.7 To improve the
spontaneous emission efficiency, Aleksanyan et al. investigated
insulator structures with one or more quantum wells in which the
quantized energy levels inside the potential wells limit the
available tunneling states to those for inelastic tunneling in
which a surface plasmon is generated..sup.8 A problem with
embodiments in the literature devices based on multiple quantum
well structures is the non-uniformity of the barrier and quantum
well layers. As another approach, Belenov et al. demonstrated
enhanced light emission in evaporated MIMIMIM triple-barrier diodes
over single-barrier diodes..sup.9 However, Belenov et al. attribute
the increased light intensity to increased scattering in the films
with additional layers. It is submitted that traditional thin film
deposition techniques, such as evaporation, are not well suited for
growing very thin (<5 nm) uniform layers of amorphous materials
due to the difficulty in controlling the uniformity of layers.
Another drawback of the Lambe device is the inefficient output
coupling of surface plasmon waves to optical radiation. In their
device, and most others in the literature, surface roughness is
used to transform the confined surface plasmon wave in the MIM
junction into radiation. This approach has the disadvantages that
the optical power must traverse the thickness of the top metal
layer before radiating into free space, which is a lossy process.
Furthermore, for some applications, surface roughening is a rather
uncontrolled method to couple out the light as the light will tend
to radiate equally into the hemisphere above the device.
Several researchers also theorized the possibility of stimulated
emission of surface plasmons in MIM diodes. For example, Siu et al.
discuss the possibility of stimulated electron tunneling in a point
contact MIM diode structure..sup.10 Siu et al. assert the existence
of stimulated electron tunneling due to surface plasmon excitation
in the point contact MIM structure as demonstrated in the negative
differential resistance exhibited in theoretical calculations and
in preliminary observations using high resistance, small area
tungsten-on-gold point contact diodes. However, Siu et al. do not
suggest practical applications for the phenomenon. As another
example of theoretical research, Drury and Ishii discuss stimulated
emission of surface plasmons in metal-insulator-metal diodes and
suggest their use as coherent optical sources..sup.11, 12 Also,
Aleksanyan et al. and Belenov et al. theorized on the amplification
of surface plasmons in MIM junctions by stimulated emission..sup.8,
13, 14, 15 Aleksanyan et al. and Belenov et al. analyzed double-
and triple-barrier structures, predicting near unity quantum
efficiency in structures with two quantum wells (triple barrier)
such that the plasmon-producing transition was due to the inelastic
tunneling between quantum levels in these two wells. In one report,
Belenov et al. discuss experimental results from a
Au--Al.sub.2O.sub.3--In.sub.2O.sub.3--Al.sub.2O.sub.3--In.sub.2O.sub.3--A-
l.sub.2O--Ag diode, showing a slight peak in the luminescence
output corresponding to a peak in the current-voltage curve (a
region of negative differential resistance)..sup.15 They do not,
however, show definitive amplification of surface plasmons. Xie
also considers stimulated emission and absorption of surface
plasmons in MIM transition metal type like junctions..sup.16 While
Xie shows that stimulated emission of surface plasmons is
theoretically possible in MIM devices, the material selection in
the MIM diode as discussed in the article is very limited. Namely,
the collector of the MIM diode in the article of Xie is required to
be a transition metal, the insulator layer must be very thin
(around 10 .ANG.), and the stimulated emission only occurs at
wavelengths longer than .lamda..about.10 .mu.m. Further, Xie
restricts the outer metal layers of the MIM to single crystal
metals. However, Xie does demonstrate the possibility of obtaining
large gain from such devices and suggests that such a structure
might be useful as an amplifier or emitter in the infrared region,
although Xie does not discuss the possibility of gain in the
visible or other spectra. Xie further suggests that the electron
tunneling can be used as a pumping mechanism in plasmon devices
with micro-dimensions, although he does not propose specific ways
in which such devices might be implemented.
None of the above cited references discuss specific device
structures, for example, for optical amplifiers or lasers, such as
the configuration of input and output optical couplers or of
resonant cavity design, but discuss only the electronic structure
of the tunneling junction. Furthermore, none of the above
references suggest methods for producing the necessary thin,
uniform barrier and well layers for these devices.
Several researchers have proposed optical modulators based on the
coupling of light with surface plasmon waves. In nearly all cases,
reflected or waveguided light is attenuated by coupling into the
surface plasmon mode, and this coupling is controlled by changing
the index of refraction of a dielectric layer adjacent to the metal
film. Thus, a fairly strong electro-optic effect is required
(.DELTA.n.about.10.sup.-3 10.sup.-2) to operate the optical
modulator. For example, Schildkraut proposed a reflection modulator
in which light internally reflected within a prism can couple to a
thin metal film on the backside of the prism by means of an
electro-optic material adjacent to the metal film..sup.17 With an
electro-optic material of second order susceptibility (e.g.,
2.times.10.sup.-7 esu), Schildkraut reports that a 100 volt signal
modulates the optical reflectance from 0.00 to 0.84. As another
example, Jannson et al. disclose a high-speed light modulator in
U.S. Pat. No. 5,067,788, which includes a buffer and an
electro-optic material sandwiched between two metal electrode
layers, positioned adjacent to and separated from an optical
waveguide (such as an optical fiber)..sup.18 A surface plasmon wave
is generated on the metal electrode adjacent to the optical
waveguide by evanescent wave coupling. Applying a voltage across
the electro-optic material changes the index of refraction of the
material and thus the strength of the surface plasmon coupling. The
optical intensity within the optical waveguide is therefore
modulated in the device of Jannson.
As another example, in U.S. Pat. No. 6,034,809 Anemogiannis
discloses an optical plasmon wave attenuator and modulator
structures in which the output optical power is attenuated or
modulated by controlling the amount of coupling between a guided
optical signal and a separately generated surface plasmon
wave..sup.19 Anemogiannis further considers the phase matching
conditions for the coupling wave vectors to achieve efficient
attenuation and modulation. Similarly, Janunts and Nerkararyan also
proposed a modulator in which light is incident on a metal film
through a prism coupled into a surface plasmon wave in the metal
and which, in turn, radiates into an optical mode in a dielectric
waveguide..sup.20 Coupling of the light into the waveguide is
controlled by an electro-optic material on top of the metal film
and dielectric waveguide. Again, large changes in index of
refraction, .DELTA.n>0.001, are required to produce
modulation.
All of the aforementioned modulators require the use of a suitable
electro-optic material, such as a non-centrosymmetric crystal or
nonlinear polymer. This requirement adds cost to device fabrication
and limits device integration with other optoelectronic devices. In
addition, these specialized materials generally require large
voltages (10 100 V) to attain the necessary refractive index change
for modulation.
The present invention provides surface plasmon devices which is
intended to reduce or eliminate the foregoing problems in a highly
advantageous and heretofore unseen way and which provides still
further advantages.
REFERENCES
.sup.1A. N. Korotkov, D. V. Averin and K. K. Likharev, "TASERs:
Possible dc pumped terahertz lasers using interwell transitions in
semiconductor heterostructures," Applied Physics Letters, Vol. 65,
No. 15, pp. 1865 1867 (1994).
.sup.2H. Drexler, J. S. Scott, S. J. Allen, K. L. Campman and A. C.
Gossard, "Photon-assisted tunneling in a resonant tunneling diode:
Stimulated emission and absorption in the THz range," Applied
Physics Letters, Vol. 67, No. 19, pp. 4102 4104 (1995).
.sup.3K. Kempa, P. Bakshi, C. G. Du, G. Feng, A. Scorupsky, G.
Strasser, C. Rauch, K. Uterrainer and E. Gornik, "Towards
stimulated generation of coherent plasmons in nanostructures,"
Journal of Applied Physics, Vol. 85, No. 7, pp. 3708 3712
(1999).
.sup.4M. Asada, Y. Oguma and N. Sashinaka, "Estimation of interwell
terahertz gain by photon-assisted tunneling measurement in
triple-barrier resonant tunneling diodes," Applied Physics Letters,
Vol. 77, No. 5, pp. 618 620 (2000).
.sup.5A. Tredicucci, C. Gmachi, M. C. Wanke, F. Capasso, A.
Hutchinson, D. L. Sivco, S.-N. G. Chu and A. Y. Cho, "Surface
plasmon quantum cascade lasers at .lamda..about.19 .mu.m," Applied
Physics Letters, Vo. 77, No. 15, pp. 2286 2288 (2000).
.sup.6J. J. Lambe and S. L. McCarthy, "Solid state source of
radiant energy having a controllable frequency spectra
characteristic," U.S. Pat. No. 4,163,920, issued Aug. 7, 1979.
.sup.7R. A. Volkov and A. F. Chuiko, "Tunneling-assisted photon
emission in MIM junctions," Physica Status Solidi (b), J63, pp. 311
320 (1991).
.sup.8A. G. Aleksanyan, E.M. Belenov, I. A. Poluektov, V. I.
Romanenko, and A. V. Uskov, "Feasibility of developing a tunable
oscillator utilizing a system of metal-barrier-metal-barrier-metal
junctions," Soviet Journal of Quantum Electronics, Vol. 11, No. 5,
pp. 635 637 (1981).
.sup.9E. M. Belenov, I. N. Kompanets, A. A. Krokhotkin, A. V.
Lezhnev, I. A. Poluektov, Yu. M. Popov, S. I. Sagitov, E. M.
Soboleva, A. G. Sobolev, A. V. Uskov, and V. G. Tsukanov,
"Investigation of the radiation emitted by metal-barrier-metal
structures," Soviet Journal of Quantum Electronics, Vol. 13, No. 4,
pp. 451 455 (1983).
.sup.10D. P. Siu, R. K. Jain, and T. K. Gustafson, "Stimulated
electron tunneling in metal-barrier-metal structures due to surface
plasmons," Applied Physics Letters, Vol. 28, No., 7, pp. 407 410
(1976).
.sup.11[D. M. Drury and T. K. Ishii, "Theory of infrared and
optical frequency amplification in metal-barrier-metal diodes, IEEE
Transactions on Microwave Theory and Techniques, Vol. MTT-27, No.
6, pp. 598 603 (June 1979).
.sup.12D. M. Drury and T. K. Ishii, "A stimulated inelastic
tunneling theory of negative differential resistance in
metal-insulator-metal diodes, IEEE Journal of Quantum Electronics,
Vol. QE-16, No. 1, pp. 58 69 (January 1980).
.sup.13E. M. Belenov, P. N. Luskinovich, V. I. Romanenko, A. G.
Sobolev, and A. V. Uskov, "Resonant tunneling in multilayer
structures in the presence of surface electromagnetic waves,"
Soviet Technical Physics Letters, Vol. 12, No. 4, pp. 200 202
(1986).
.sup.14E. M. Belenov, P. N. Luskinovich, V. I. Romanenko, A. G.
Sobolev, and A. V. Uskov, "Emission of surface electromagnetic
waves in the case of resonance tunneling of electrons," Soviet
Journal of Quantum Electronics, Vol. 17, No. 10, pp. 1348
1352(1987).
.sup.15E. M. Belenov, I. N. Kompanets, Yu.M. Popov, I. A.
Poluektov, S. I. Sagitov, E. M. Soboleva, A. G. Sobolev, A. V>
Uskov, and V. G. Tsukanov, "Amplification of surface plasma
oscillations in complex metal-barrier-metal structures," Soviet
Journal of Quantum Electronics, Vol. 12, No. 7, pp. 930 931
(1982).
.sup.16L. Z. Xie, "Stimulated emission of surface plasmons in
metal-insulator-metal (transition metal type like) structures,"
Infrared Physics, Vol. 25, No. 5, pp. 661 664 (1985).
.sup.17.Jay S. Schildkraut, "Long-range surface plasmon
electrooptic modulator," Applied Optics, Vol. 27, No. 21, pp. 4587
4590 (1988).
.sup.18T. P. Jannson, J. L. Jannson and B. Moslehi, "High
modulation rate optical plasmon waveguide modulator," U.S. Pat. No.
5,067,788, issued Nov. 26, 1991.
.sup.19E. Anemogiannis, "Optical plasmon-wave structures," U.S.
Pat. No. 6,034,809, issued Mar. 7, 2000.
.sup.20N. A. Janunts and Kh.V. Nerkararyan, "Modulation of light
radiation during input into waveguide by resonance excitation of
surface plasmons," Applied Physics Letters, Vol. 79, No. 3, pp. 299
301 (2001).
SUMMARY OF THE INVENTION
As will be described in more detail hereinafter, there is disclosed
herein a device designed in accordance with one aspect of the
present invention and including an input port, which input port is
configured to receive an input signal, and an output port. The
device also includes a structure, which structure in turn includes
a tunneling junction connected with the input port and the output
port. The tunneling junction is configured in a way (i) which
provides electrons in a particular energy state within the
structure, (ii) which produces surface plasmons in response to the
input signal, (iii) which causes the structure to act as a
waveguide for directing at least a portion of the surface plasmons
along a predetermined path toward the output port such that the
surface plasmons so directed interact with the electrons in a
particular way, and (iv) which produces at the output port an
output signal resulting from the particular interaction between the
electrons and the surface plasmons.
In another aspect of the invention, an associated method of
providing a device, as described above, is disclosed. The method
includes the steps of providing an input port configured to receive
an input signal and providing an output port. The method further
includes the steps of providing a structure including a tunneling
junction and connecting the tunneling junction with the input port
and the output port, the tunneling junction (i) providing electrons
in a particular state within the structure, (ii) causing the
structure to produce surface plasmons in response to the input
signal, (iii) causing the structure to act as a waveguide for
directing at least a portion of the surface plasmons along a
predetermined path toward the output port such that the surface
plasmons so directed interact with the electrons in a particular
way, and (iv) producing at the output port an output signal
resulting from and in response to the particular interaction
between the electrons and the surface plasmons.
In still another aspect of the invention, a detector for detecting
input electromagnetic radiation incident thereon is disclosed. The
detector includes an input port, which input port is configured to
receive the input electromagnetic radiation, and an output port.
The detector of the present invention further includes a structure,
which structure in turn includes a tunneling junction connected
with the input port and the output port. The tunneling junction is
configured in a way (i) which provides electrons in a particular
energy state within the structure, (ii) which produces surface
plasmons in response to the input electromagnetic radiation, (iii)
which causes the structure to act as a waveguide for directing at
least a portion of the surface plasmons along a predetermined path
toward the output port such that the surface plasmons so directed
interact with the electrons in a particular way, and (iv) which
produces at the output port an output electrical signal resulting
from the particular interaction between the electrons and the
surface plasmons.
In yet another aspect of the invention, an amplifier for amplifying
input electromagnetic radiation incident thereon is disclosed. The
amplifier includes an input port, which input port is configured to
receive the input electromagnetic radiation, and an output port.
The amplifier further includes a structure, which structure in turn
includes a tunneling junction connected with the input port and the
output port. The tunneling junction is configured in a way (i)
which provides electrons in a particular energy state within the
structure, (ii) which produces surface plasmons in response to the
input electromagnetic radiation, (iii) which causes the structure
to act as a waveguide for directing at least a portion of the
surface plasmons along a predetermined path toward the output port
such that the surface plasmons so directed interact with the
electrons in a particular way, and (iv) which produces an output
electromagnetic radiation with a given gain at the output port, the
output electromagnetic radiation resulting from the particular
interaction between the electrons and the surface plasmons.
In a still further aspect of the invention, an emitter is
disclosed. The emitter of the present invention includes an input
port, which input port is configured to receive an input electrical
current, and an output port. The emitter further includes a
structure, which structure in turn includes a tunneling junction
connected with the input port and the output port. The tunneling
junction includes first and second non-insulating layers and an
arrangement disposed between the first and second non-insulating
layers. The arrangement is configured to serve as a transport of
electrons between the first and second non-insulating layers and
includes at least two insulating layers. The tunneling junction is
configured in a way (i) which causes the electrons to be in a
particular energy state within the structure, (ii) which produces
surface plasmons in response to the input electrical current, (iii)
which causes the structure to act as a waveguide for directing at
least a portion of the surface plasmons along a predetermined path
toward the output port, and (iv) which produces at the output port
an output electromagnetic radiation resulting from the particular
interaction between the electrons and the surface plasmons.
In another aspect of the invention, a laser is disclosed. The laser
of the present invention includes an input port, which input port
is configured to receive an input electrical current, and an output
port. The laser further includes a structure, which structure in
turn includes a tunneling junction connected with the input port
and the output port. The tunneling junction is configured in a way
(i) which provides electrons in a particular energy state within
the structure and (ii) which produces surface plasmons in response
to the input electrical current. Still further, the laser includes
a reflective assembly configured to form a resonant cavity
surrounding the structure. The reflective assembly is configured
for selectively confining at least a portion of the surface
plasmons within the resonant cavity while selectively directing
another portion of the surface plasmons along a predetermined path
toward the output port such that the surface plasmons so confined
interact with the electrons in a particular way so as to produce an
output electromagnetic radiation, which output electromagnetic
radiation results from the particular interaction between the
electrons and the portion of surface plasmons confined within the
resonant cavity.
In yet another aspect of the invention, a modulator is disclosed.
The modulator of the present invention includes an input port,
which input port is configured to receive an input electromagnetic
radiation, and an output port. The modulator further includes a
signal source for providing a modulation signal. Still further, the
modulator includes a structure, which structure in turn includes a
tunneling junction connected with the input port and the output
port. The tunneling junction is configured for receiving the
modulation signal from the signal source. The tunneling junction is
further configured in a way (i) which provides electrons in a
particular energy state within the structure responsive to and as a
function of the modulation signal, (ii) which produces surface
plasmons in response to the input electromagnetic radiation, (iii)
which causes the structure to act as a waveguide for directing at
least a portion of the surface plasmons along a predetermined path
toward the output port such that the surface plasmons so directed
interact with the electrons in a particular way, and (iv) which
produces at the output port an output electromagnetic radiation
resulting from the particular interaction between the electrons and
the surface plasmons such that the output electromagnetic radiation
displays at least one characteristic that is a function of the
modulation signal.
In a yet further aspect of the invention, a filter is disclosed.
The filter of the present invention includes an input port, which
input port is configured to receive an input electromagnetic
radiation over a range of frequencies, and an output port. The
filter further includes a structure including an insulating layer
and a non-insulating layer disposed adjacent to the insulating
layer such that an interface is defined between the insulating and
non-insulating layers. The insulating and non-insulating layers are
connected with the input port and the port, and are configured in a
way which produces surface plasmons in response to the input
electromagnetic radiation at the interface between the insulating
and the non-insulating layers. Still further, the filter includes a
reflective assembly configured to form a resonant cavity
surrounding the structure. The reflective assembly is configured
for selectively confining at least a portion of the surface
plasmons within the resonant cavity while selectively directing
another portion of the surface plasmons along a predetermined path
toward the output port so as to produce an output electromagnetic
radiation having predetermined frequencies selected from within the
range of frequencies.
In a still further aspect of the invention, a polarization rotator
is disclosed. The polarization rotator of the present invention
includes an input port, which input port is configured to receive
an input electromagnetic radiation having a first polarization
state, and an output port. The polarization rotator further
includes a main body connected with the input port and the output
port. The main body includes a waveguide structure and a
non-insulator arrangement at least partially surrounding the
waveguide structure. The main body is configured for producing
surface plasmons having the first polarization state in response to
the input electromagnetic radiation. The waveguide structure and
the non-insulator arrangement are configured for directing at least
a portion of the surface plasmons along a predetermined path toward
the output port in a way which produces at the output port an
output electromagnetic radiation having a second polarization
state.
DESCRIPTION OF THE DRAWINGS
The present invention may be understood by reference to the
following detailed description taken in conjunction with the
drawings briefly described below.
FIG. 1A is a diagrammatic illustration of a top view of a detector
including a traveling wave detection region designed in accordance
with the present invention, shown here to illustrate an
implementation including an input antenna configuration.
FIG. 1B is a diagrammatic illustration in partial cross section of
the detector of FIG. 1A, shown here to illustrate the layered
structure of the traveling wave detection region of the
detector.
FIG. 1C is a diagrammatic illustration in partial cross section of
an alternative configuration of the detector of FIG. 1A, shown here
to illustrate an alternative layered structure of the traveling
wave detection region of the detector.
FIG. 1D is a diagrammatic illustration in partial cross section of
another alternative configuration of the detector of FIG. 1A, shown
here to illustrate another alternative layered structure of the
traveling wave detection region of the detector.
FIG. 2A is a diagrammatic illustration of a side view of still
another alternative configuration for the detector designed in
accordance with the present invention, shown here to illustrate an
alternative input coupler including a grating.
FIG. 2B is a diagrammatic illustration of a side view of yet
another alternative configuration for the detector designed in
accordance with the present invention, shown here to illustrate
another alternative input coupler including an evanescent wave
coupler.
FIG. 2C is a diagrammatic illustration of a side view of a further
alternative configuration for the detector designed in accordance
with the present invention, shown here to illustrate an output
coupler configuration including an antireflection arrangement.
FIG. 3A is a diagrammatic illustration of a side view of a detector
including a gain region designed in accordance with the present
invention.
FIG. 3B is a diagrammatic illustration of a side view of still
another alternative configuration of the detector designed in
accordance with the present invention, shown here to illustrate the
details of a tuning arrangement.
FIG. 4A is a diagrammatic illustration of a top view of an
amplifier including a traveling wave gain region designed in
accordance with the present invention, shown here to illustrate an
implementation including an input antenna and an output
antenna.
FIG. 4B is a diagrammatic illustration of a side view of an
alternative configuration for the amplifier designed in accordance
with the present invention, shown here to illustrate alternative
input and output couplers including gratings.
FIG. 4C is a diagrammatic illustration of a side view of another
alternative configuration for the amplifier designed in accordance
with the present invention, shown here to illustrate alternative
input and output couplers including evanescent wave couplers.
FIG. 5A is a diagrammatic illustration of a top view of an emitter
including a tunneling/plasmon waveguide region designed in
accordance with the present invention, shown here to illustrate an
implementation including an output antenna configuration.
FIG. 5B is a diagrammatic illustration in partial cross section of
the emitter of FIG. 5A, shown here to illustrate the layered
structure of the traveling wave detection region of the
detector.
FIG. 5C is a diagrammatic illustration of a side view of an
alternative configuration for the emitter designed in accordance
with the present invention, shown here to illustrate an alternative
output coupler including a grating.
FIG. 5D is a diagrammatic illustration of a side view of another
alternative configuration for the emitter designed in accordance
with the present invention, shown here to illustrate another
alternative output coupler including an evanescent wave
coupler.
FIG. 6A is a diagrammatic illustration of a side view of a laser
including a plasmon wave resonator designed in accordance with the
present invention, shown here to illustrate an implementation
including an output grating coupler.
FIG. 6B is a diagrammatic illustration of a side view of an
alternative configuration for the laser designed in accordance with
the present invention, shown here to illustrate an alternative
resonator cavity configuration including distributed Bragg
reflectors and an alternative output coupler including an
evanescent wave coupler.
FIG. 7A is a diagrammatic illustration of a side view of another
alternative configuration for the laser designed in accordance with
the present invention, shown here to illustrate an alternative
plasmon wave resonator configuration including gain regions at the
cavity nodes of the resonator.
FIG. 7B is a diagrammatic illustration of a side view of yet
another alternative configuration for the laser designed in
accordance with the present invention, shown here to illustrate
another alternative plasmon wave resonator configuration including
gain regions at the cavity nodes of the resonator.
FIG. 8A is a diagrammatic illustration of a side view of still
another alternative configuration for the laser designed in
accordance with the present invention, shown here to illustrate an
alternative output coupler including a gap coupler
configuration.
FIG. 8B is a diagrammatic illustration of a side view of another
alternative configuration for the laser, which is designed in
accordance with the present invention and includes a gap coupler
into a MIM waveguide.
FIG. 8C is a diagrammatic illustration of a side view of yet
another alternative configuration for the laser designed in
accordance with the present invention, shown here to illustrate
another alternative output coupler including a leaky reflector.
FIG. 8D is a diagrammatic illustration of a side view of still
another alternative configuration for the laser, which is designed
in accordance with the present invention and includes an impedance
mismatched antenna as the output coupler.
FIG. 9A is a diagrammatic illustration in perspective view of a
ring laser designed in accordance with the present invention.
FIG. 9B is a diagrammatic illustration in perspective view of an
alternative configuration for the ring laser designed in accordance
with the present invention, shown here to illustrate an alternative
output coupling configuration.
FIG. 9C is a diagrammatic illustration in perspective view of yet
another alternative configuration for the ring laser designed in
accordance with the present invention, shown here to illustrate an
electrooptic layer configuration to provide a tunable ring
laser.
FIG. 9D is a diagrammatic illustration in perspective view of
another alternative configuration for the ring laser, which is
designed in accordance with the present invention and includes an
annular electrooptic arrangement to provide another tunable ring
laser.
FIG. 10 is a diagrammatic illustration of a top view of a modulator
designed in accordance with the present invention, shown here to
illustrate an implementation including input and output antennae
configurations.
FIG. 11 is a diagrammatic illustration of a side view of a
configuration for a modulator with a gain region designed in
accordance with the present invention, shown here to illustrate an
implementation including input and output grating couplers.
FIG. 12 is a diagrammatic illustration of a side view of a filter
designed in accordance with the present invention, shown here to
illustrate an implementation including end reflectors and input and
output grating couplers.
FIG. 13 is a diagrammatic illustration of a side view of a tunable
filter designed in accordance with the present invention, shown
here to illustrate an implementation including distributed Bragg
reflectors and input and output evanescent wave couplers.
FIG. 14 is a diagrammatic illustration of a side view of an
alternative configuration for the tunable filter designed in
accordance with the present invention, shown here to illustrate an
implementation including end reflectors, input and output grating
couplers and a tunneling junction region.
FIG. 15 is a diagrammatic illustration of a side view of a tunable
filter with gain designed in accordance with the present invention,
shown here to illustrate an implementation further including a
tuning region and a gain region with input and output grating
couplers.
FIG. 16 is a diagrammatic illustration in perspective view of a
waveguide polarization device designed in accordance with the
present invention, shown here to illustrate an implementation
including an optical fiber and a shaped, metal strip.
FIG. 17 is a diagrammatic illustration of a top view of a
polarizing beam splitter designed in accordance with the present
invention, shown here to illustrate an implementation including
slab waveguides.
FIG. 18 is a diagrammatic illustration in perspective view of a
slow mode-fast mode plasmon wave converter designed in accordance
with the present invention, shown here to illustrate an
implementation on a slab substrate.
DETAILED DESCRIPTION
The following description is presented to enable one of ordinary
skill in the art to make and use the invention and is provided in
the context of a patent application and its requirements. Various
modifications to the described embodiments will be readily apparent
to those skilled in the art and the generic principles herein may
be applied to other embodiments. Thus, the present invention is not
intended to be limited to the embodiment shown but is to be
accorded the widest scope consistent with the principles and
features described herein.
Although the aforedescribed prior art devices and theoretical
disclosures regarding sophisticated manipulation of surface
plasmons, such as stimulated emission and stimulated absorption,
have been made over the past twenty-five years, the Applicants are
unaware of presently available, practical devices that actually
allow such manipulations. Recent progress in tunneling junction
technology by the assignee of the present application has greatly
increased the flexibility in fabrication and design of electron
tunneling devices based on metal-insulator(s)-metal structures,
thus allowing the fabrication of high speed electron tunneling
devices (See, copending U.S. patent application Ser. No. 09/860,988
and copending U.S. patent application Ser. No. 09/860,972, both of
which applications are incorporated herein by reference). However,
the aforementioned prior art devices as well as the devices of the
copending U.S. patent applications Ser. Nos. 09/860,988 and
09/860,972 are based on small area tunneling junctions (i.e., the
tunneling junction region is on the order of square microns or less
in area) and therefore tend to be RC-limited, and these lumped
small area devices do not provide a waveguide path to guide the
propagation of surface plasmons through the devices. For example,
although the device of Lambe et al. provides broadband emission
from a tunneling junction structure, the surface plasmon
propagation is limited to random emission through the top metal
layer, which has been etched so as to be slightly porous.
Therefore, it would be desirable to provide a predetermined
waveguide path along which surface plasmons can be directed toward,
for example, an output port or a separate functional unit
(detector, modulator, amplifier, etc.), and, optionally, to provide
a particular path along which an arrangement may be provided to
tune the operation of the device.
Turning now to the drawings, wherein like components are indicated
by like reference numbers throughout the various figures, attention
is immediately directed to FIGS. 1A 1D. In FIG. 1A, a detector 10
designed in accordance with the present invention is illustrated.
Detector 10 includes an input antenna (enclosed by dashed lines
12), which is designed to receive input electromagnetic radiation
13 incident thereon. Input antenna 12 is formed of a first metal
layer 14 and a second metal layer 16. Input antenna 12 is
configured to convert the received input electromagnetic radiation
into surface plasmons (not shown) and to direct them by waveguide
action toward a detection region (enclosed by dashed lines 20),
which is integrally formed from first metal layer 14 and second
metal layer 16. Detection region 20 is configured to absorb the
surface plasmons via rectification and, consequently, generate an
output electrical signal which is directed to an output 22. Output
22 may be, for example, an ammeter or oscilloscope for measuring
and viewing the output electrical signal.
The details of detection region 20 are illustrated in FIG. 1B,
which shows a cross-sectional view of a portion of detector 10 of
FIG. 1A. As can be seen in FIG. 1B, detection region 20 is formed
of first metal layer 14 and second metal layer 16 with an
insulating arrangement 24 disposed therebetween so as to define a
tunneling junction structure while providing a waveguide for the
surface plasmons generated by the input antenna. In accordance with
one aspect of the invention, the input signal from the input
antenna enters from only one end of the tunneling junction which
naturally produces traveling waves in the tunneling junction. This
contrasts with prior art devices in which the input signal enters
from multiple sides (opposite, adjacent, or other different sides)
of the tunneling junction.
For purposes of the present application, a tunneling junction
structure is understood to be a structure including two outer
layers of non-insulative material(s) spaced apart such that an
insulative arrangement, such as one more layers of a dielectric
material, is positioned therebetween. Examples of such tunneling
junction structures are disclosed in the aforementioned co-assigned
and copending U.S. patent applications Ser. Nos. 09/860,988 and
09/860,972, both of which are incorporated herein by reference. In
particular, in one embodiment of the present invention, first metal
layer 14 is formed of a 40 nm thick layer of niobium deposited by
sputtering. Insulating arrangement 24 may be a native oxide of
first metal layer 14 and a second oxide layer, such as tantalum
oxide, deposited by atomic layer deposition (ALD) or formed by
sputtering. Alternatively, both insulator layers may be deposited
by ALD or formed by sputtering. Second metal layer 16 in this
embodiment can be a 40 nm thick layer of tantalum deposited by
sputtering methods.
Continuing to refer to FIG. 1B, the first metal layer 14,
insulating arrangement 24 and second metal layer 16 are of
predetermined widths to enable guiding of surface plasmons received
from the input antenna toward the output. When the surface
plasmons, generated by the input antenna in response to the input
electromagnetic radiation, reach detection region 20, some of the
surface plasmons are absorbed with a given absorption coefficient
by electrons present in the detection region such that at least
some of the electrons are caused to tunnel between first and second
metal layers 14 and 16. The tunneling electrons result in the
generation of an electrical current between first and second metal
layers 14 and 16, which electrical current is directed toward the
output. In other words, the surface plasmons interact with
tunneling electrons in the detection region such that rectification
of the energy of the surface plasmons in detection region 20
results in the generation of the output electrical signal.
Particularly, the output electrical signal is produced by surface
plasmon-induced electron tunneling through the detection
region.
Still referring to FIGS. 1A and 1B, the elongated configuration of
detection region 20 serves to guide the surface plasmons as a
traveling, slow mode surface plasmon wave so as to eliminate the
time constant limitation concerns of the aforementioned lumped
small area devices. Since the slow-mode surface plasmon wave
propagates along the tunneling junction, as opposed to
perpendicularly through the tunneling junction as in the case of
the lumped small area devices, elongated, detection region 20 does
not have the same limitations on the RC time constant as the lumped
small area devices. Lumped small area devices operate by creating a
uniform oscillating field across the tunneling junction, and
therefore RC limitations apply. In the traveling wave devices,
however, the oscillating field across the tunneling junction also
propagates along the tunneling junction, hence RC time constant is
not a limitation and essentially only affects the propagation
velocity of the traveling surface plasmon wave. In addition, the
configuration of detection region 20 makes impedance matching of
detection region 20 to input antenna 12 easier than in small area
devices. In a lumped small area device, the impedance of the
optical antenna element must be matched to the RC equivalent
circuit impedance of the tunneling junction, which can be thousands
or millions of ohms for tunneling junctions with areas of
10.sup.-10 cm.sup.2. This impedance matching consideration can pose
difficulties for the antenna designer. In the traveling wave
detector, by comparison, the waveguide impedance for the slow mode
surface plasmon is on the order of 40 50 ohms (for example, for 0.8
eV light), which impedance value is much easier to match with an
antenna or optical coupler. Furthermore, the impedance of the
lumped small area devices varies exponentially with the thickness
of the insulating layer, and hence impedance matching to a specific
antenna impedance is difficult. The traveling wave devices have a
characteristic impedance that is much less dependent on the
thickness of the insulating layer, therefore the processing window
that provides for good matching to the antenna is much wider, thus
increasing fabrication yield. Therefore, the width of detection
region 20 should be designed so as to efficiently guide the surface
plasmon waves through the device toward the output along a
predetermined path.
Referring now to FIGS. 1C and 1D in conjunction with FIGS. 1A and
1B, the design of detection region 20 can be optimized, for
example, for low-pass filtering by making the waveguide area large,
which increases the RC time constant and lowers the maximum signal
bandwidth detectable (while the optical carrier frequency response
remains high). Such a solution would help minimize noise in
detectors designed to operate at low signal frequencies.
Furthermore, different layer designs are possible to tailor the
electron tunneling characteristics through the tunneling junction.
For example, insulating arrangement 24 can include a single
dielectric layer, multiple dielectric layers, a combination of
dielectric and metal layers, semiconductors and/or superconductors.
For instance, as shown in FIG. 1C, insulating arrangement 24 can
include first and second dielectric layers 32 and 34 positioned
adjacent to each other to enhance the nonlinearity of the electron
tunneling process therethrough, as disclosed in detail in the
aforementioned co-assigned and copending United States Patent
Applications. Also, as shown in FIG. 1D, insulating arrangement 24
can include first and second dielectric layers 42 and 44 as well as
a third metal or insulator quantum well layer 46 to form a
double-barrier resonant tunneling diode.
Although input antenna 12 is shown as bowtie antenna in FIG. 1A,
other antenna and input coupler designs are possible as long as
proper impedance matching is achieved between the input coupler and
the detection region. In FIGS. 2A and 2B, two alternative input
coupler configurations are shown. FIG. 2A illustrates an embodiment
of a detector 100 formed on a substrate 110. Detector 100 includes
an input grating coupler (enclosed by dashed lines 112) formed on
top of a first metal layer 114. A detection region (enclosed by
dashed lines 120) is formed by first metal layer 114 and a second
metal layer 116 with an insulating arrangement 124 disposed
therebetween. Insulating arrangement 124 can be, for example, one
of the layer configurations illustrated in FIGS. 1B 1D. Detector
100 receives input TM-polarized electromagnetic radiation 13 at
input grating coupler 112, which then converts the received input
electromagnetic radiation into surface plasmons and directs the
surface plasmons into detection region 120.
Alternatively, in FIG. 2B, a detector 150 including an input
evanescent wave coupling region 152 is shown. Electromagnetic
waveguide region is separated from first metal layer 114 by a
barrier layer 156 and also formed on a substrate 158. Detector 150
is configured to receive input electromagnetic radiation 13 via an
electromagnetic waveguide region 154. The input TM-polarized
electromagnetic radiation is coupled into the device as a surface
plasmon wave and directed into detection region 120 to be rectified
in the manner described above. First metal layer 114 and barrier
layer 156 should be thin enough such that the input electromagnetic
radiation can be coupled into detection region 120 as a surface
plasmon wave. Surface plasmon penetration depth into metals is on
the order of 20 30 nm for infrared radiation so that metal layer
114 should be 20 30 nm or slightly thinner to maximize field
penetration.
In addition, it is known that tunneling junctions, such as in the
elongated waveguide structure of the detector of the present
invention, including two outer metal layers with an insulating
arrangement (whether a single insulating layer or multiple layers
combining and insulating and non-insulating layers) generally
supports three fundamental surface plasmon traveling wave modes:
the aforedescribed slow mode, which interacts strongly with
tunneling electrons but has a relative short propagation length due
to high loss, and two fast modes, which do not interact so strongly
with tunneling electrons and have relatively long propagation
lengths. It is possible to minimize slow mode surface plasmon wave
reflections between the input coupler and the detection region by
adding a quarterwave antireflection layer between these two
portions of the detector such that strong interaction with
tunneling electrons is maintained in the detector. A key aspect of
the antireflection layer is that its insulator thickness is chosen
to null out back-reflections between the layers of thick insulator
and very thin, tunnel insulator.
As an example, FIG. 2C illustrates a detector 160, which is a
modification of detector 150 of FIG. 2B and includes an
antireflection arrangement, indicated by double-headed arrow 162.
Antireflection arrangement 162 is disposed between detection region
120 and input evanescent wave coupling region 152. Antireflection
arrangement 162 includes an additional metal layer 164 and an
additional insulator layer 166. The thickness of additional
insulator layer 166, indicated by double-headed arrow 168, is such
that the slow mode surface plasmon propagation constant in that
region is equal to the geometric mean of the propagation constants
in the two surrounding regions. The width of antireflection
arrangement 162, corresponding to the length of double-headed arrow
162, is chosen to be a quarter wave of the electromagnetic
radiation wavelength for which detector 160 is designed.
FIG. 3A illustrates yet another embodiment of a detector 200
designed in accordance with the present invention. Detector 200 is
similar to detector 100 of FIG. 2A and includes input grating
coupler 112. Like detector 100 of FIG. 2A, detector 200 includes
first metal layer 114 deposited along the top of substrate 110.
However, detector 200 includes a second metal layer 216 and a first
insulating arrangement 224 forming a detection region (indicated by
a double headed arrow 220), which detection region 220 is separated
from input grating coupler 112 by a gain region (indicated by
another double headed arrow 230). Gain region 230 is also a
tunneling junction structure formed of a third metal layer 232 and
a second insulating arrangement 234. While detection region 220,
like detection region 20 of FIG. 1A, is designed for rectification
of surface plasmons to provide an output electrical signal at
output 22, gain region 230 serves as a plasmon wave "preamplifier"
to provide additional surface plasmons by a process of stimulated
emission. That is, gain region 230 is configured such that, when it
receives the surface plasmons generated by input grating coupler
112, the surface plasmons so received cause plasmon-induced
electron tunneling in the gain region. Unlike detection region 220,
gain region 230 is further configured such that the tunneling
electrons give up their energies to stimulate the emission of
additional plasmons in the gain region. All of the surface plasmons
are then directed into the detection region for rectification and
subsequent generation of the output electrical signal.
Still further, detector 150 of FIG. 2B can be made into a tunable
device. As shown in FIG. 3B, an electrical arrangement 252 is
provided to supply an electrical voltage between first metal layer
114 and electrode 258, thus across an electrooptic material 260
positioned adjacent to the metal layer of the plasmon coupling
region 152. Varying the voltage provided by electrical arrangement
252 changes the coupling condition from the dielectric waveguide
154 into the metal layer 114 and thereby shifts the optical
frequency which will couple into the detector. As in detector 150,
detection region 120 includes insulating arrangement 124 disposed
between first and second metal layers 114 and 116 such that
detection region 120 forms a tunneling junction structure for
rectification of surface plasmons directed therein.
Turning now to FIGS. 4A 4C, an amplifier designed in accordance
with the present invention is described. FIG. 4A illustrates an
amplifier 300 including an input antenna (enclosed by dashed lines
312), which input antenna is configured to receive input
electromagnetic radiation 313. Input antenna 312 is formed of a
combination of one end of a first metal layer 314 and one end of a
second metal layer 316. The opposing ends of first metal layer 314
and second metal layer 316 form an output antenna (enclosed by
dashed lines 318). First and second metal layers 314 and 316
further serve to define a gain region (enclosed by another set of
dashed lines 320) between input antenna 312 and output antenna 318.
Gain region 320 can include, for example, an insulator arrangement
(such as multiple dielectric layers, quantum dots, etc., not shown)
as described in the foregoing discussion of the detector of the
present invention such that gain region 320 serves as a tunneling
junction structure. Amplifier 300 is configured such that input
antenna 312 receives input electromagnetic radiation 313. Input
antenna 312 then converts the received input electromagnetic energy
into surface plasmons and directs the surface plasmons toward gain
region 320. As described above with regard to gain region 230 shown
in FIG. 3A, gain region 320 of amplifier 300 is designed such that
the surface plasmons from input antenna 312 cause electrons to
tunnel between first and second metal layers 314 and 316, thereby
producing additional surface plasmons by stimulated emission. Then,
the additional surface plasmons are directed into output antenna
318, which in turn converts at least a portion of the surface
plasmons directed therein into output electromagnetic radiation
322.
Gain region 320 is configured to maximize the stimulated emission
process so as to provide maximum gain while minimizing loss. For
example, by proper design of the layers of the tunneling junction
structure, the efficiency of the stimulated emission process can be
enhanced. By appropriate tunneling junction design, the probability
of stimulated emission exceeds the probability of stimulated
absorption in order to help minimize surface plasmon loss during
propagation through the gain region. For instance, gain region 320
can include a layered structure such as, but not limited to, one or
more dielectric layers, one or more additional metal layers and/or
a discontinuous layer, such as a layer of quantum dots. As an
example, gain region 320 can include a triple-barrier (i.e., two
quantum well) tunneling structure in which the stimulated emission
transition occurs between quantum levels in the two well regions.
Alternatively, the quantum levels could also be realized, for
example, by forming isolated metal quantum dots by incomplete
coalescence of a thin metal layer. The resonant tunneling through
these quantum well and quantum dot structures yields highly
nonlinear current-voltage response, such as negative differential
resistance, and, subsequently, high stimulated emission
efficiency.
The amplifier of the present invention is particularly advantageous
in that the device is not RC limited due to the traveling wave
design. Therefore, the tunneling junction can be made larger than
in point contact devices, thus improving manufacturing tolerances
and reducing the device susceptibility to thermal damage and static
or transient charge damage. While these traveling wave devices are
larger than point contact tunnel junction devices, they are
considerably smaller than equivalent optoelectronic devices built
on dielectric waveguides, such as semiconductor optical amplifiers.
This compactness is due to the strong guiding nature of surface
plasmons compared to optical waves in dielectric waveguides. As a
result, while a single mode dielectric waveguide may have cross
sectional dimensions of several microns or even tens of microns, a
slow mode surface plasmon waveguide may have cross sectional
dimensions of only a hundred to a thousand nanometers. This device
compactness is a great advantage particularly for integrated optics
applications.
Other loss minimization considerations can also be made. As
mentioned above with respect to detectors, tunneling junctions
generally support one slow mode and two fast modes of surface
plasmon traveling waves. Therefore, the gain region can be designed
to maximize the propagation efficiency of the slow mode surface
plasmons in order to maximize the gain while minimizing loss. For
example, the tunneling junction thickness can be optimized. It is
known that thicker junctions have lower tunneling current, and
hence, lower gain per unit length; however thicker tunneling
junctions also have lower slow mode surface plasmon wave
propagation loss since more of the optical field drops across the
insulator and not the metal layers. Therefore, the tunneling
junction thickness can be optimized to balance the gain per length
and slow mode surface plasmon wave propagation loss. Moreover,
since slow mode surface plasmon propagation loss is predominantly
due to conductive loss in the metal layers, the selection of highly
reflective metals, such as the noble metals gold (Au), silver (Ag),
aluminum (Al) and copper (Cu), can minimize the propagation loss.
Other metals, such as niobium (Nb) and tantalum (Ta), may also be
good choices because their reflectivity values are reasonably high
in the infrared and their native oxides so as to result in low
barrier height and therefore low resistance tunneling junctions.
Also, as discussed in relation to the aforedescribed devices of the
present invention, an antireflection layer can be incorporated into
the amplifier between the input antenna and the gain region and/or
between the gain region and the output antenna to reduce coupling
loss between the antennae and the gain region.
Minimization of loss in the device can also lead to a reduction in
the noise level of the device. Decreased loss would lead to lower
gain thresholds, which lowers the probability of spontaneous
emission of surface plasmons in the device. As a result, the noise
level in the amplifier would be lowered.
As in the case of the aforedescribed detector, other input/output
coupler designs are possible for use with the amplifier of the
present invention. For example, FIG. 4B illustrates an amplifier
400 including an input grating coupler 412 and an output grating
coupler 418 in place of the input and output antennae shown in FIG.
4A. Also, in FIG. 4C, an amplifier 450 including an input
evanescent wave coupling region (indicated by a double headed arrow
452) and an output evanescent wave coupling region (indicated by
another double headed arrow 456) in place of input and output
antennae 312 and 318 of FIG. 4A is shown.
Furthermore, an active feedback mechanism (not shown) can be added
to the amplifier of the present invention. Due to the sensitivity
of surface plasmons to small changes in properties of the
metal-dielectric interface, small temperature perturbations can
cause relatively large fluctuations in the surface plasmon
wavelength. This effect can be problematic, for example, for
resonant structures such as laser cavities. An active feedback
mechanism, such as a voltage-controlled electro-optic layer beside
the insulating layer, can be used to compensate for such
temperature fluctuations. The feedback arrangement can be
configured to monitor the output of the amplifier and, if the
output signal is reduced by a certain amount, to apply the
appropriate tuning voltage across tunneling region such that the
amplifier provides a substantially constant gain. For example,
digital signal restoration can be provided as follows. By limiting
the rate at which charge is supplied to the electrodes of the gain
medium, the gain will saturate at a particular surface plasmon
amplitude. One way to accomplish this is to supply current to 416
through a resistor, such that when the surface plasmons draw a
current above a predetermined level, then the voltage applied to
the electrodes drops below the level required to provide additional
gain.
Attention is now directed to FIGS. 5A 5D, illustrating embodiments
of an emitter designed in accordance with the present invention.
First considering FIG. 5A, an emitter 500 includes an input current
source 510 and an output antenna (enclosed in dashed lines 512).
Output antenna 512 is formed of a first metal layer 514 and a
second metal layer 516. First metal layer 514 and second metal
layer 516 are extended to integrally form a surface plasmon
waveguide section (indicated by another set of dashed lines 520).
Surface plasmon waveguide section 520 is configured to receive an
input current from input current source 510 such that the input
current is applied across first metal layer 514 and second metal
layer 516. The input current causes electrons to tunnel between the
first and second metal layers so as to generate surface plasmons in
surface plasmon waveguide section 520 by spontaneous emission of
incoherent surface plasmons. In other words, the input current
imparts energy to at least some of the electrons in the surface
plasmon waveguide section. Then, the energized electrons
spontaneously lose their energy and, consequently, spontaneously
emit surface plasmons. Surface plasmon waveguide section 520 is
further configured to direct the surface plasmons so generated
toward output antenna 512 to be coupled out as broadband,
incoherent, output electromagnetic radiation 522. In essence,
emitter 500 works like a light emitting diode (LED) by using an
electrical current as an input to generate a broadband, incoherent
electromagnetic output by spontaneous emission. That is, emitter
500 takes electrical current as the input and generates a broadband
electromagnetic radiation as the output, which is a function of the
input.
In accordance with one aspect of the invention, the output signal
exits from only one end of the tunneling junction. This contrasts
with prior art devices in which the output signal exits from
multiple locations on the tunneling junction.
FIG. 5B shows a cross sectional view of surface plasmon waveguide
section 520 of FIG. 5A. Surface plasmon waveguide section 520
includes an insulating arrangement 524 positioned between first
metal layer 514 and second metal layer 516. In the embodiment shown
in FIG. 5B, insulating arrangement 524 in turn includes a first
dielectric layer 532 directly adjacent to a second dielectric layer
534. Alternatively, insulating arrangement 524 can further include
additional components such as, but not limited to, one or more
additional metal layers, one or more additional dielectric layers
and/or a discontinuous layer, such as a layer of quantum dots (not
shown). The layered structure design of surface plasmon waveguide
section 520 should be configured to maximize the spontaneous
emission of surface plasmons within the surface plasmon waveguide
section. For example, double-barrier (one quantum well) and
triple-barrier (two quantum well) structures can increase the
spontaneous emission probability over single barrier designs by
increasing the probability of inelastic tunneling and reducing the
probability of elastic tunneling. In addition, to further enhance
the coupling of the generated surface plasmons out of output
antenna 512, an antireflection layer as described above can be
positioned between surface plasmon waveguide section 520 and output
antenna 512.
FIGS. 5C and 5D illustrate other output coupler options other than
the output antenna shown in FIG. 5A for the emitter of the present
invention. FIG. 5C shows an emitter 600 including an output grating
coupler 618, indicated by dashed lines, for converting the surface
plasmons generated in a surface plasmon waveguide region 620,
enclosed by another set of dashed lines, into output
electromagnetic radiation 622. Output grating coupler 618 works
essentially the same way as output coupler 418 of FIG. 4B. Also, in
FIG. 5D, an emitter 650 includes an electromagnetic waveguide
region 652 with an output evanescent wave coupling region
(indicated by a double headed arrow 654) for coupling out the
surface plasmons generated in surface plasmon waveguide region 620.
Output evanescent wave coupling region 654 functions in
substantially the same way as output evanescent wave coupling
region 456 as shown in FIG. 4C.
As a modification to the emitter embodiments illustrated in FIGS.
5A 5D, the surface plasmon waveguide regions can be designed to
provide gain so as to generate not only surface plasmons by
spontaneous emission but also additional surface plasmons by
stimulated emission. Such a surface plasmon waveguide region can be
implemented by appropriate design of the insulating arrangement
disposed between the first and second metal layers. For example, a
layered design such as the one used in gain region 230 of the
detector with gain shown in FIG. 3 can be used in insulating
arrangement 524 or 624, with a modification that insulating
arrangements 524 and 624 are further configured to maximize
spontaneous emission of surface plasmons.
Turning now to FIG. 6A, a laser 700 designed in accordance with the
present invention is described. Laser 700 is based on a substrate
710 and also includes an input current source 712. Like in the
aforedescribed devices of the present invention, a first metal
layer 714 and a second metal layer 716 serve as the outer metal
layers of a tunneling junction structure. An insulating arrangement
724 is disposed between first metal layer 714 and second metal
layer 716 such that a gain region (indicated by a set of dashed
lines 720) is formed, as in the aforedescribed amplifier and the
detector with gain. Gain region 720 further includes a first
reflector 726 and a second reflector 728 defining a resonant cavity
therebetween. Gain region 720 is configured to receive input
current from input current source 712 such that tunneling electrons
are caused to tunnel between first metal layer 714 and second metal
layer 716. As in the case of the aforedescribed amplifier as shown
in FIGS. 4A 4C, the tunneling electrons in turn generate coherent,
surface plasmons within the resonant cavity by stimulated emission
so as to provide gain. Second reflector 728 is designed to be
partially transmissive such that a portion of the surface plasmons
thus generated are directed to an output grating coupler (indicated
by another set of dashed lines 718) and coupled out of laser 700 as
output electromagnetic radiation 722. Since the additional surface
plasmons are generated by stimulated emission, the resulting output
electromagnetic radiation is coherent.
The linewidth of output electromagnetic radiation 722 can be made
very narrow by designing the gain region to minimize the
probability of spontaneous emission occurring therein. For example,
insulating arrangement 724 can include one or more dielectric
layers, one or more additional metal layers or one or more
discontinuous metal layers, such as quantum dots. Additionally, by
using the double- or triple-barrier tunneling structure, for
instance, the gain region of the laser can be designed to maximize
the probability of stimulated emission of additional surface
plasmons while minimizing the probability of stimulated absorption
of the surface plasmons within the tunneling junction structure.
The linewidth is also a function of the resonant cavity length and
the resonant cavity Q-factor, which can both be appropriately
controlled in the design and fabrication processes. Since the
slow-mode surface plasmon within gain region 720 has a much higher
propagation constant than the fast-mode surface plasmon, the
reflectivity of second reflector 728 should be quite high, thus
increasing the Q-factor of the resonant cavity. Also, by altering
the input current, the wavelength and/or intensity of the output
electromagnetic radiation can be tuned. Furthermore, an active
feedback arrangement, as described above, can be added to enhance
the output stability of the device. An example design of the
tunneling junction structure in a laser is a triple-barrier, double
quantum well structure using, for instance, tantalum oxide or
aluminum oxide for the barrier layers and niobium oxide or thin
metal layers for the quantum wells. The width of the laser cavity
should be chosen (for example, a halfwave of the design wavelength)
so as to null out any laterally propagating plasmons.
A modification to the laser of FIG. 6A is shown in FIG. 6B. A laser
750 includes an electromagnetic waveguide region 752. Output
grating coupler 718 of laser 700 of FIG. 6A is replaced with an
output evanescent wave coupling region (indicated by double headed
arrow 754), which output evanescent wave coupling region is
configured to couple the surface plasmons directed thereon into
electromagnetic waveguide region 752 as output electromagnetic
radiation 722. Output evanescent wave coupling region 754 functions
in substantially the same way as output evanescent wave coupling
region 456 as shown in FIG. 4C. As in output evanescent coupling
region 456, first metal layer 714 should be thin in order for the
surface plasmon at the top interface of the first metal layer to
couple into waveguide region 752 below. Furthermore, the resonant
cavity in gain region 760 is defined by a first distributed Bragg
reflector 762 and a second distributed Bragg reflector 764. First
and second distributed Bragg reflectors 762 and 764 are configured
to reflect surface plasmons of a predetermined frequency at which
the gain region is designed to provide gain.
FIG. 7A illustrates another modification to the laser shown in FIG.
6A. A laser 800 is generally the same in configuration as laser 700
of FIG. 6A. However, in laser 800, a gain region (enclosed by
dashed lines 820) includes a plurality of metal segments 816 such
that a tunneling junction structure is formed only between each
metal segment 816 and first metal layer 714 and, as a result, gain
is provided only between each metal segment 816 and first metal
layer 714. Since the resonant cavity defined by first and second
reflectors 726 and 728 has specific regions therein where nodes of
the optical field are located, the position of metal segments 816
is designed such that gain is efficiently provided at those cavity
nodes and not at anti-nodes where the thin tunneling junctions
would only contribute to optical loss.
FIG. 7B shows an alternative embodiment of the laser of FIG. 7A
with gain provided at cavity nodes only. Laser 800B is
substantially similar in configuration to laser 800 of FIG. 7B, but
metal segments 816 of laser 800 are connected together into a
segmented electrode 816B in laser 800B. Segmented electrode 816B is
configured such that a plurality of insulating blocks 824, which
are integrally formed from insulating arrangement 724B, fill in
between the conducting segments in the segmented electrode. In this
way, gain is provided at the cavity nodes of the resonant cavity. A
further advantage of the segmented electrode configuration as shown
in FIG. 7B is that the slow mode surface plasmon propagation of the
traveling wave is maintained while loss is reduced due to the
thicker tunneling junction. Still further, the regions defined by
insulating blocks 824 can also be designed to act as resonant
layers.
Still other possible modifications to the laser of the present
invention are illustrated in FIGS. 8A and 8B. In FIG. 8A, a laser
900 is similar in structure to laser 700 of FIG. 6A with the same
gain region and resonant cavity configuration. However, instead of
an output grating coupler, laser 900A includes a gap coupler
configuration including a waveguide structure 902 separated from
gain region 720 by a gap (indicated by a double headed arrow 904).
Waveguide structure 902 includes barrier layers 905 and 906
surrounding a dielectric waveguide 907. A portion of the surface
plasmons from the resonant cavity then couples across the gap and
into waveguide structure 902 as output electromagnetic radiation
922. The width of gap 904 determines the amount of output
electromagnetic radiation that is coupled out of the resonant
cavity and the gap width may be, for example, a quarterwave or less
of the design wavelength.
Alternatively, the gap coupler configuration can include a MIM slow
mode surface plasmon waveguide, as shown in FIG. 8B. A laser 900B
includes a MIM waveguide 930 separated from gain region 720 by a
gap (indicated by a double headed arrow 904B). MIM waveguide 930
includes a thick insulating layer 932 topped by an electrode 934.
The thick insulating layer 932 helps to reduce loss of slow mode
surface plasmon waves during propagation through the MIM
waveguide.
As yet another alternative, a laser 950 shown in FIG. 8C is similar
to laser 750 of FIG. 6B. However, the second distributed Bragg
reflector 764 of laser 750 is replaced with a leaky distributed
Bragg reflector (indicated by dashed lines 952), which leaky
distributed Bragg reflector 952 allows a predetermined amount of
the surface plasmons within the resonant cavity to be coupled out
to a waveguide section 954 provided adjacent to the resonant
cavity. In this case, the "leakiness" or of the amount of
transmission allowed through the leaky reflector determines the
amount of the surface plasmons that is coupled out of the resonant
cavity. The "leakiness" of the reflector can be controlled, for
example, by the number of alternating dielectric layers used in the
dielectric stack forming the leaky reflector. That is, a reflector
with fewer dielectric layers will result in higher
transmission.
Alternatively, an impedance-mismatched output antenna can be used
as the partially transmissive reflector such that a predetermined
amount of the surface plasmons within the resonant cavity is
coupled out of the antenna as the output electromagnetic radiation.
For example, as shown in FIG. 8D, a laser 970 includes a gain
region 972 disposed between an input reflector 974 and an impedance
mismatched antenna (indicated by dashed lines 976). Impedance
mismatched antenna 976 is formed from a first metal layer 978 and a
second metal layer 980, which also form gain region 972. The
intentional impedance mismatch between gain region 972 and the
antenna structure provides high reflection of slow mode surface
plasmon waves back into gain region 972, thus effectively forming a
resonant cavity surrounding gain region 972. The impedance mismatch
should also be designed such that a certain amount of surface
plasmon waves are allowed to couple out through impedance
mismatched antenna 976 in order for laser 970 to provide output
electromagnetic radiation 982.
As another alternative embodiment of the laser of the present
invention, a ring laser 1000 designed in accordance with the
present invention is illustrated in FIG. 9A. An input current is
provided by an input current source 1012 across a first metal layer
1014 and a second metal layer 1016 with an insulating arrangement
1018 disposed therebetween. First metal layer 1014, second metal
layer 1016 and insulating arrangement 1018 are configured to form a
circular, tunneling junction structure such that the input current
causes tunneling electrons to tunnel between the first and second
metal layers and, consequently, to generate surface plasmons within
the circular, tunneling junction. The circular, tunneling junction
structure is configured such that the tunneling junction itself
serves to confine the generated surface plasmons therein such that
additional surface plasmons are generated by stimulated emission.
The output signal can be coupled out, for example, by evanescent
coupling by adding an MIM slow mode waveguide 1020 adjacent to, but
slightly separated from, the edge of the ring laser. Waveguide 1020
includes, for example, an insulator layer 1022 sandwiched between
metal layers 1024 and 1026. Waveguide 1020 is separated from ring
laser 1000 by a gap (indicated by double-headed arrow 1028) such
that light from ring laser 1000 is coupled into waveguide 1020 by
evanescent wave coupling.
Alternatively, for instance, a V-shaped antenna can be made to
spiral out of the edge of the ring laser. For example, as shown in
FIG. 9B, ring laser 1030 includes a first metal layer 1032 and a
second metal layer 1034 with an insulating arrangement (not shown)
disposed therebetween. A plurality of extensions 1036 and 1038 are
integrally formed from first metal layer 1032 and second metal
layer 1034, respectively, so as to form a plurality of V-shaped
antennae. In this case, light incident on the ring laser would be
funneled into the ring laser and/or emitted light would be coupled
out of the ring laser.
As another alternative scheme for coupling light out of the ring
laser of the present invention, one of the surfaces of the ring
laser configuration (for example, first metal layer 1014 of ring
laser 1000 as shown in FIG. 9A) can be made rough or patterned with
a grating such that light would be emitted from the roughened or
patterned surface. It would also be possible to have an edge
emitting ring laser by appropriate finishing of the ring edge.
A possible modification to the aforedescribed ring laser is to make
it into a tunable ring laser. As illustrated in FIG. 9C, one way to
achieve such a tunable ring laser 1040 is to make sure first metal
layer 1014 is no thicker than approximately a skin depth at the
optical frequency (.about.30 nm) so that the some of the surface
plasmon electric field penetrates the electrode. Tunable ring laser
1040 also includes an electrooptic material 1042 on top of first
metal layer 1014. Furthermore, a third metal layer 1044 is
deposited on top of electrooptic material 1042 such that a voltage
can be supplied from a voltage source 1046 across electrooptic
material 1042 (i.e., between first metal layer 1014 and third metal
layer 1044). The variable voltage thus supplied modulates the
electrooptic properties of the electrooptic material such that the
resonant wavelength of the surface plasmon wave in the ring laser
is modulated, thus tuning the output wavelength from the tunable
ring laser. Alternatively, as shown in FIG. 9D, the outer edge of
the tunneling junction structure, which forms the ring laser, can
be coated with an electrooptic material and an annular electrode
layer. A tunable ring laser 1050 includes a circular tunneling
junction structure formed by first metal layer 1014 and second
metal layer 1016 with an insulating arrangement 1018 disposed
therebetween such that input current can be applied between the
first and second metal layers by input current source 1012. Tunable
ring laser 1050 further includes an electrooptic material 1052 and
a ring electrode 1054 surrounding the circular tunneling junction.
A voltage source 1056 supplies a voltage between the ring electrode
and first metal layer 1014, a portion of the surface plasmon field
that penetrates into electrooptic material 1052 will produce a
shift in the surface plasmon wavelength and hence shift the
resonant frequency of tunable ring laser 1050.
Turning now to FIG. 10, a modulator 1100 designed in accordance
with the present invention is shown. Modulator 1100 includes an
input antenna (1112 enclosed in dashed lines 1112), which input
antenna 1112 is formed at one end of a first metal layer 1114 and a
second metal layer 1116. Input antenna 1112 is configured to
receive input electromagnetic radiation 1102 and to convert the
input electromagnetic radiation into surface plasmons. The opposing
end of first metal layer 1114 and second metal layer 1116 form an
output antenna (enclosed by another set of dashed lines 1118).
First metal layer 1114 and second metal layer 1116 further form a
tunneling junction structure (indicated by dashed lines 1120)
between the input and output antennae. Tunneling junction structure
1120 is configured receive the surface plasmons from input antenna
1112 and to serve as a waveguide to direct the surface plasmons
toward output antenna 1118 while absorbing a predetermined amount
of the surface plasmons with a given absorption coefficient, as in
the case of the aforedescribed detector shown in FIG. 1A.
Furthermore, tunneling junction structure 1120 is configured to
receive a modulation signal 1122 from a modulation input source
1124. Still further, tunneling junction 1120 is configured such
that the absorption coefficient is altered in response to
modulation signal 1122 so as to allow different amounts of the
received surface plasmons from the input antenna to reach the
output antenna depending on the modulation signal. The surface
plasmons that reach output antenna 1118 are then coupled out of
modulator 1100 as modulated output electromagnetic radiation 1126
such that the modulated output electromagnetic radiation displays
at least one characteristic of modulation signal 1122, such as, for
example, amplitude modulation. Amplitude modulation can be realized
in this device, for example, by using modulation signal 1122 as a
variable bias voltage across tunneling junction 1120 such that the
probability of stimulated absorption (is modulated. If the
insulator layer in the tunneling junction is made sufficiently
thick, amplitude modulation can be decreased while increasing the
amount of phase modulation. Thus, the nonlinearity in tunneling
current induced by the surface plasmon effectively modifies the
propagation constant of the slow mode surface plasmon wave through
second and higher order nonlinearity values in a manner similar to
the linear electro-optic effect in non-centrosymmetric
crystals.
A possible modification to the aforedescribed modulator is shown in
FIG. 11. A modulator 1200 of FIG. 11 is configured for modulating
input electromagnetic radiation 1202 and includes a substrate 1210
on which a first metal layer 1214 is deposited. A second metal
layer 1216 and a first insulating arrangement 1218, along with
first metal layer 1214, together form a tunneling junction
(enclosed in a set of dashed lines 1220). Tunneling junction 1220
is configured to receive a modulation signal 1222 from a modulation
input source 1224 such that the absorption coefficient of tunneling
junction 1220 is modulated in response to modulation signal 1222.
Additionally, a third metal layer 1226 and a second insulating
arrangement 1224, along with first metal layer 1214, together form
a gain region (enclosed in dashed lines 1230) to serve as a surface
plasmon "preamplifier" before tunneling junction 1220. Modulator
1200 further includes an input grating coupler (indicated by dashed
lines 1240), which input grating coupler is configured to receive
input electromagnetic radiation 1202 and to convert the input
electromagnetic radiation so received into surface plasmons to be
directed into gain region 1230. The surface plasmons cause
tunneling electrons to tunnel between first metal layer 1214 and
third metal layer 1226 such that additional surface plasmons are
produced in gain region 1230 by stimulated emission. The surface
plasmons so produced are then directed into tunneling junction
1220, which absorbs different amounts of the surface plasmons
depending on modulation signal 1222 while allowing the remainder of
surface plasmons to reach an output grating coupler 1242, indicated
by another set of dashed lines. Output grating coupler 1242 then
couples the remaining surface plasmons out of modulator 1200 as
modulated output electromagnetic radiation 1244. In this way,
modulator 1200 provides gain as well as modulation in the output
electromagnetic radiation 1244.
Still other modifications are possible to the embodiments of the
modulator of the present invention. Aforedescribed antireflection
layers can be positioned between the input couplers and the
tunneling junction and/or the tunneling junction and the output
couplers to minimize the surface plasmon losses during propagation
through the device. The insulating arrangement in the tunneling
junction can be designed to maximize the modulation in the
absorption coefficient induced by the modulation signal. For
example, the insulating arrangement can include multiple layers of
dielectrics and/or metals to enhance this feature. That is, the
higher the nonlinearity in the tunneling junction current-voltage
curve, the more efficient the modulation will be. Thus, resonant
tunneling structures, such as those described in the aforementioned
co-assigned and copending patent applications and those of double-
and triple-barrier diodes, would yield the highest modulation
efficiency--highest change in optical properties at lowest change
in voltage/current. Other output couplers, such as the evanescent
wave coupler described above, and combinations of different output
couplers can also be used in the modulator.
As a possible modification to the aforedescribed modulator, it is
often desirable to increase the degree of phase modulation produced
relative to the degree of amplitude modulation to form a modulator
that has low insertion loss. One way to accomplish this is to
increase the thickness of the insulating layer above that of an
efficient detector, such that the tunneling probability across it
is decreased. The result is that the loss from amplitude modulation
is reduced, allowing the phase modulation to dominate.
Turning now to FIG. 12, a filter 1300 designed in accordance with
the present invention is shown as an example of a surface plasmon
device which does not require a tunneling junction structure.
Filter 1300 is essentially a resonant cavity surface plasmon filter
which is designed to filter the desired frequencies out of an input
electromagnetic radiation 1302 having a range of input frequencies.
Filter 1300 includes a substrate 1310 on which a first metal layer
1314 has been deposited. Filter 1300 further includes a first
reflector 1315 and a second reflector 1317 positioned at opposing
facets of an insulating arrangement 1318. Insulating arrangement
1318 can be, for example, a layer of dielectric material or even
air. First reflector 1315 and second reflector 1317 define a
resonant cavity across insulating arrangement 1318. First metal
layer 1314, first and second reflectors 1315 and 1317 and
insulating arrangement 1318 together form a filtering region
(enclosed in dashed lines 1320). Filter 1300 also includes an input
grating coupler 1322, which is configured to receive input
electromagnetic radiation 1302 and to convert the electromagnetic
radiation so received into surface plasmons to be directed into
filtering region 1320. The filtering region is configured like a
Fabry-Perot resonant cavity to only transmit predetermined
frequencies of surface plasmons waves such that only surface
plasmons of those predetermined frequencies are directed into an
output grating coupler 1324 to be coupled out as output
electromagnetic radiation 1326.
Various modifications may be made in the filter of the present
invention. For example, a tunable filter 1400 shown in FIG. 13 is
designed to filter input electromagnetic radiation 1402 received
via an electromagnetic waveguide region 1410, which is surrounded
by barrier region 1411 and 1413. A first metal layer 1414 is
disposed on top of electromagnetic waveguide region 1410. A tuning
input source 1412 is configured to supply a tuning input (e.g., a
variable electrical voltage or current, not shown) via an electrode
1416 across an insulating arrangement 1418 so as to tune a material
property of the insulating arrangement depending on the tuning
input. For example, the tuning input, such as a variable electrical
voltage applied across the insulating arrangement, can cause a
change in the refractive index of an electro-optic material 1418. A
filtering region (enclosed in dashed lines 1420) includes a portion
of first metal layer 1414, insulating arrangement 1418, with
electrode 1416 thereon, a first distributed Bragg reflector 1422
and a second distributed Bragg reflector 1424. First and second
distributed Bragg reflectors 1422 and 1424, positioned on opposite
sides of insulating arrangement 1418, together define a resonant
cavity. Therefore, by applying the tuning input across insulating
arrangement 1418, the effective path length of the resonant cavity
is altered and, as a result, the wavelength and hence resonant
frequency of the surface plasmons allowed through the resonant
cavity is changed.
As another modification to the filter of the present invention, a
tunneling junction structure can be incorporated into the resonant
cavity to provide gain to the surface plasmons within the resonant
cavity such that the intensity of the output electromagnetic
radiation is increased. For example, a filter 1500 shown in FIG. 14
is configured to filter input electromagnetic radiation 1502 and
includes a substrate 1510 with a first metal layer 1514 deposited
thereon. Filter 1500 further includes a second metal layer 1516 and
an insulating arrangement 1518 which, together with first metal
layer 1514, form a tunneling junction (enclosed by dashed lines
1520). In this case, insulating arrangement 1518 is configured so
as to provide additional surface plasmons within tunneling junction
1520 by stimulated emission, as in the case of the aforedescribed
amplifier of the present invention. Tunneling junction 1520 also
includes a first reflector 1522 and a second reflector 1524
defining a resonant cavity. In this way, output electromagnetic
radiation 1530 coupled out of filter 1500 will be stronger in
intensity than if the tunneling junction were not present. In other
words, filter 1500 acts essentially as an tuned amplifier.
As still another modification, an electro-optic material and the
tunneling junction structure can both be incorporated into the
resonant cavity structure simultaneously so as to provide a tunable
amplifier. Such a filter 1600, which is configured to filter input
electromagnetic radiation 1602, is shown in FIG. 15. Filter 1600
includes a substrate with a first metal layer 1614 positioned
thereon. A second metal layer 1616 and an insulating arrangement
1618, together with first metal layer 1614, define a tunneling
junction 1620, indicated by a double headed arrow, which tunneling
junction is configured to provide gain as described above. Filter
1600 further includes an electro-optic material 1622 with an
electrode 1624 disposed thereon to form a tuning block 1626,
indicated by another double headed arrow. Tuning block 1626 is
configured to receive a tuning input (such as an electrical
voltage) from a tuning input source 1628 such that an electro-optic
property of electro-optic material 1622 is altered by the tuning
input. A first reflector 1630 and a second reflector 1632 are
positioned on opposing ends of the combination of the tuning block
and the tunneling junction so as to define a resonant cavity.
Therefore, by applying the tuning input to electro-optic material
1622, the resonance condition within the resonant cavity can be
changed, thus altering the frequency of surface plasmons allowed
therethrough.
As another device based on the control of surface plasmons in
conjunction with a waveguide structure in accordance with the
present invention, a waveguide polarization device 1700 is shown in
FIG. 16. The waveguide polarization device of the present invention
takes advantage of the fact that surface plasmon generation and
propagation is polarization sensitive. Waveguide polarization
device 1700 includes an optical fiber 1702, which in turn includes
a cladding 1704 surrounding a fiber core 1706. A metal strip 1720
is arranged around the optical fiber in a helical fashion (i.e.,
the metal strip winds around the optical fiber by a quarter turn,
in the embodiment shown in FIG. 16). When an input electromagnetic
radiation 1730, including s- and p-polarization states, is
propagated through the optical fiber and encounters the metal
strip, the initially p-polarized component of the input
electromagnetic radiation propagates along metal strip 1720 as a
surface plasmon wave while the initially s-polarized component
travels through the optical fiber substantially unaffected. Due to
the orientation of metal strip 1720 around the optical fiber, the
surface plasmon wave propagating along metal strip 1720 is rotated
in polarization such that, at the end of the metal strip, the
surface plasmon wave is converted to an s-polarized electromagnetic
radiation, which then becomes combined with the initially
s-polarized component to produce an s-polarized, output
electromagnetic radiation 1740. That is, at the end of the metal
strip, the output electromagnetic radiation is s-polarized. Thus,
by proper configuration of the metal strip, the waveguide
polarization device of the present invention can serve, for
example, as a polarization combiner (if the metal strip rotates the
polarization of the p-polarized component by 90-degrees as
described above while leaving the s-polarized component unchanged)
or to provide phase delay (if the metal strip rotates the
polarization of the p-polarized component by 180-degrees; i.e., the
metal strip winds helically around the optical fiber by a half
turn, rather than a quarter turn).
Another device which takes advantage of the polarization-dependence
of the surface plasmon wave generation is a polarizing beam
splitter as shown in FIG. 17. Polarizing beam splitter 1800
includes a substrate 1810 on which first and second dielectric
waveguide sections 1820 and 1822 are arranged. First and second
dielectric waveguide sections 1820 and 1822 are connected by a
metal strip 1830 leading from first dielectric waveguide section
1820 to second dielectric waveguide section 1822. An input
electromagnetic radiation 1840, including s- and p-polarization
states, is coupled into the first dielectric waveguide section.
When input electromagnetic radiation 1840 encounters metal strip
1830, the initially p-polarized component of the input
electromagnetic radiation propagates along the metal strip as a
surface plasmon wave while the initially s-polarized component
travels through first dielectric waveguide section 1820 essentially
unaffected. In this way, the initially s-polarized component of
input electromagnetic radiation 1840 is directed through first
dielectric waveguide section 1820 as a first output electromagnetic
radiation 1850 with s-polarization, while the initially p-polarized
component is directed through second dielectric waveguide section
1822 as a second output electromagnetic radiation 1852 with
p-polarization. Thus, the initially s- and p-components of input
electromagnetic radiation 1840 has been split into two beams, each
having only s- or p-polarized electromagnetic radiation.
Turning now to FIG. 18, a slow mode-fast mode converter 1900 is
disclosed. As discussed in the foregoing description of the devices
of the present invention, one of the considerations in considering
surface plasmon propagation through a device is the fact that three
modes of surface plasmon waves (i.e., two fast modes and one slow
mode) are simultaneously generated. Although the slow mode
interacts more strongly with tunneling electrons in a tunneling
junction structure, the slow mode also is more lossy. The fast
modes tend to interact less strongly with electrons but have longer
propagation lengths. It is sometimes desirable to be able to
convert from the slow mode to the fast mode and vice versa. In
general, the difference between slow mode and fast mode surface
plasmon waves in a tunneling junction structure is that, in slow
mode surface plasmons, the electromagnetic field component
propagating along the interface between the bottom metal layer and
the middle insulating layer(s) is out of phase with the
electromagnetic field component traveling along the interface
between the middle insulating layer(s) and the top metal layer. In
fast mode surface plasmon waves, the electromagnetic field
components traveling along the two interfaces are in phase.
Therefore, in order to convert from the fast mode to the slow mode
and vice versa, it is necessary to manipulate the relative phase of
the electromagnetic field components of the surface plasmon waves
at the different interfaces.
Slow mode-fast mode converter 1900, shown in perspective in FIG.
18, includes a substrate on which a first metal strip 1920 and a
second metal strip 1922 has been deposited, with an input port 1930
defined at one end of the metal strips and an output port 1940
defined at an opposing end of the metal strips. The metal strips
are separated by an insulating arrangement 1924. As shown in FIG.
18, second metal strip 1922 includes an extra curved segment not
found in first metal strip 1920. It should be noted that the curved
segment forms a continuous curve rather than any abrupt turn or
change in direction, thus preventing possible reflections at abrupt
interfaces. Therefore, while the electromagnetic field component of
a surface plasmon wave traveling along first metal strip 1920
travels straight across substrate 1910, the electromagnetic field
component of the surface plasmon wave traveling along second metal
strip 1922 is forced to travel an extra distance along the
semicircular segment. By proper selection of the length of the
semicircular segment, a 180-degree phase delay is introduced in the
electromagnetic field component traveling along second metal strip
1922 relative to the electromagnetic field component traveling
along first metal strip 1920 such that surface plasmon wave which
is in the slow mode at an input port 1930 will be converted from a
slow mode to one of the fast modes (and vice versa) at output port
1940. That is, slow mode-fast mode converter 1900 converts from the
fast mode to the slow mode and vice versa by configuring the top
metal layer to introduce an extra phase delay in the
electromagnetic field component traveling along the top metal layer
compared to the electromagnetic field component traveling along the
bottom metal layer.
Various modifications to the converter as shown in FIG. 18 can be
contemplated. For example, antireflection layers can be positioned
between the input port and the semicircular segment and/or the
semicircular segment and the output port to reduce coupling losses
at the input and/or output. Also, the shape of the semicircular
segment can be modified according to fabrication and performance
requirements.
It is especially notable that the aforedescribed devices of the
present invention are based on the concepts of the conversion of an
input signal (electrical or electromagnetic) into surface plasmons
within the devices, manipulation of the surface plasmon waves so
generated within the devices, and the subsequent conversion of the
manipulated, surface plasmons into an output signal (electrical or
electromagnetic). That is, unlike, for example, the devices of
Anemogiannis which merely modulate the evanescent wave components
of input electromagnetic waves, the devices of the present
invention converts essentially all of the energy of the input
signal into surface plasmons so as to manipulate the energy of the
surface plasmons themselves in achieving effects such as stimulated
or spontaneous emission of additional surface plasmons and
stimulated absorption of the surface plasmons. As a result, the
modulation or signal frequency response may be RC limited, but the
optical response of the aforedescribed devices of the present
invention are not limited by RC considerations. Furthermore, the
devices of the present invention include predetermined waveguide
paths to guide the surface plasmon waves through the devices so as
to provide sufficient interaction lengths to manipulate the surface
plasmons in desired ways. Still further, the devices of the present
invention are flexible in design such that they can be made very
compact. Therefore, a plurality of the devices of the present
invention can be combined on a single substrate. For example, a
plurality of detectors can be formed on a single substrate to form
an array, or a number of different devices can be combined to form
an integrated optical structure.
Although each of the aforedescribed embodiments have been
illustrated with various components having particular respective
orientations, it should be understood that the present invention
may take on a variety of specific configurations with the various
components being located in a wide variety of positions and mutual
orientations and still remain within the spirit and scope of the
present invention. Furthermore, suitable equivalents may be used in
place of or in addition to the various components, the function and
use of such substitute or additional components being held to be
familiar to those skilled in the art and are therefore regarded as
falling within the scope of the present invention. For example, the
aforedescribed devices can be combined with conventional optical
elements, such as , but not limited to, lasers, detectors, optical
fiber, lenses and photonic bandgap crystals. Various combinations
of materials, such as superconductors, semiconductors and
semimetals, can be incorporated into the aforedescribed devices in
order to achieve desired performance and fabrication
specifications. Two or more of the aforedescribed configurations of
the present invention can be combined to achieve multi-purpose
devices, such as, for example, a filtering detector. Atomic layer
deposition (ALD) of the various thin well and barrier layers allows
precise control of layer thickness and uniformity at the monolayer
level, thus providing significant improvement in performance over
previously reported tunneling junction devices (e.g., the devices
of Aleksanyan et al. and Belenov et al. which are based on
evaporated layers). The resonant tunneling effect through tunneling
junction structures formed by ALD yields highly nonlinear
current-voltage response and, subsequently, high efficiency.
Furthermore, the smooth layers and interfaces formed by ALD can
significantly reduce scattering loss of slow mode surface plasmon
waves, which have relatively short wavelengths (e.g., approximately
200 nm for 200 THz radiation in Al--Al.sub.2O.sub.3--Ag tunnel
junction structures). Therefore, the present examples are to be
considered as illustrative and not restrictive, and the invention
is not to be limited to the details given herein but may be
modified within the scope of the appended claims.
* * * * *